sign in sign up
<<
Home , Carrying capacity, Predators (soundtrack), Predator satiation, Surplus killing

University of Nebraska - Lincoln DigitalCommons@University of Nebraska - Lincoln

USDA National Wildlife Research Center - Staff U.S. Department of : Animal and Publications Plant Health Inspection Service

September 2001

THE ROLE OF IN WILDLIFE DYNAMICS

Eric M. Gese Utah State University, [email protected]

Frederick F. Knowlton Utah State University, Logan, UT

Follow this and additional works at: https://digitalcommons.unl.edu/icwdm_usdanwrc

Part of the Environmental Sciences Commons

Gese, Eric M. and Knowlton, Frederick F., "THE ROLE OF PREDATION IN WILDLIFE " (2001). USDA National Wildlife Research Center - Staff Publications. 542. https://digitalcommons.unl.edu/icwdm_usdanwrc/542

This Article is brought to you for free and open access by the U.S. Department of Agriculture: Animal and Plant Health Inspection Service at DigitalCommons@University of Nebraska - Lincoln. It has been accepted for inclusion in USDA National Wildlife Research Center - Staff Publications by an authorized administrator of DigitalCommons@University of Nebraska - Lincoln. THE ROLE OF PREDATIOS IN WILDLIFE POPULATION DYNAMICS

ERIC ?vI. GESE, National Wildlife Research Center, Department of and Wildlife, Utah State University, Logan, UT 84322-5295.

FREDERICK F. WOWLTON, National Wildlife Research Center, Department of Fisheries and Wildlife, Utah State University, Logan, UT 84322-5295.

Abstract: The role predation plays in the dynamics of prey is controversial. Our understandings of -prey relationships is complicated by a multitude of factors in the environment and a general lack of knowledge of most ecological systems. Various other factors, besides predation, may regulate or limit prey populations, and various factors influence the degree to which predation affects prey populations. Furthermore, some factors may create time lags, or even cause generational effects, that go unnoticed. Herein, we review the role of predation in wildlife population dynamics, some of the factors influencing predator-prey interactions, and attempt to indicate where the professional debate currently is focused and where it may need to go to enhance our understanding of predator-prey interactions.

Predation has been defined as on ungulate density. Assessments of the individuals of one eating living importance that (Canis lupus) individuals of another species (Taylor predation plays in regulating or limiting 1984). The role of predators in the moose (Alces alces) populations varies population dynamics of prey species has among . Interactions among been investigated for decades, yet moose, forage, and climate have been determining whether or not predators limit postulated to determine moose density or regulate a prey population remains (Peterson et al. 1984, Mech et al. 1987, controversial within the scientific Thompson and Peterson 1988). Bergerud et profession (e.g., Erlinge et al. 1984, Kidd al. (1983), Bergemd and Snider (1988), and and Lewis 1987, Newsome et al. 1989, Van Ballenberghe and Ballard (1994) Sinclair 1989, Sinclair et al. 1990, Messier considered predation a major 1991, 1994, Skogland 1991, Boutin 1992, of moose because moose density was Pech et al. 1992). Much of the debate generally below forage canying capacity. results from the multitude of competing Messier and CrSte (1985) and Messier variables, including predation, that influence (1991) argued that moose and predator demographics of prey species and the interactions were complex and that the difficulty of conducting large-scale, long- effect of predation varied from density- term studies with some degree of control or dependent to inversely density-dependent replication. In a review of studies involving over the range of moose densities resulting predation on ungulates, Connolly (1978) in population cycles, multiple stable states, reported that 45 studies suggested predation and predator pits. Skozland (1991) was a limiting factor, while 27 studies suggested that ths existin: data was indicarsd predation \vas not a limitin: factor inconclusive with rsgaids to predation

Gese, E. M., and F. F. Knowlton. 2001. The role of predation in wildlife population dynamics. Pages 7-25 in The Role of Predator Control as a Tool in Game Management. Edited by T. F. Ginnetr and S. E. Henke. Texas Agricultural Research and Extension Center. San Angelo. Texas. regulating moose, while Boutin (1992) 'attack, capture, or consume' commonly argued that wolf predation as a limiting found within the predation literature. The factor on moose populations was not relationship between the kill rate by a supported. predator and prey density is termed the "functional response." Lotka (1925) and In this paper, we will attempt to Volterra (1926) provided initial provide a foundation on predator-prey mathematical descriptions of predator-prey theory, describe some studies illustrating interactions, which assumed that the number the roles that predators and other variables of prey captured increased in direct can play in the dynamics of wildlife proportion to the number of predators. populations (mainly from the camivore- Nicholson and Bailey (1935) proposed the ungulate literature), and suggest reasons relationship was curvilinear with the kill rate why the debate over the influence predators decreasing as predator satiation sets an have on prey populations continues. It is upper limit to food consumption. not our intent to critically review all Subsequently, Holling (1959) described 3 predator-prey studies, but to use certain types of functional responses (Fig. 1). A studies to illustrate aspects of predator-prey Type I functional response occurs when the relationships. kill rate per predator is directly proportional to prey density. In the Type I1 response, the TERMINOLOGY kill rate is limited at higher prey densities by satiation of the predator, and is thus For our discussions of predator-prey curvilinear. The Type 111 functional relations to be fruitful, we need to clarify response is sigmoid in shape with a lag in some terminology. We also should kill rate at low prey density due to low recognize that many of the initial terms and efficiency or absence of a search early theories concerning the role predators image and an upper limit set by predator play in limiting or regulating prey satiation. Type I1 functional responses have populations were developed by been documented between and entomologists examining relationships moose (Messier 1994; Fig. 2), wolves and between numbers of parasites needed to caribou (Rangifer tarandzrs) (Dale et al. regulate invertebrate pest species on 1994), as well as (Canis latrans) agricultural crops. The terms regulating and and black-tailed jackrabbits (Lepus limiting have often been used calforniczrs) (Stoddart et al. 2001). In interchangeably, with regulation defined as addition to functional responses, Morris et "any density-dependent process that tends to al. (1958) demonstrated a "numerical stabilize population numbers over time. The response" in which predator numbers process that causes the change(s) in increase in response to increasing prey is termed limitation" . This numerical response may (Skogland 1991). We consider a limiting be from reproduction or immigation. factor to be any mortality factor that reduces Numerical responses of coyotes to changes the rate of (Ballard et al. in black-tailed jackrabbit (Kno\vlron and 2001). We also will try to adhere to using Stoddart 1992) and snowshoe hare (L. the term 'kill' to denote the essential uniericnnla) abundance (O'Donoghue et a[. component of a predator's impact upon 1997) have been documented- Messier prey, rather than the ambiguous terms (1991) found a numerical response of wolves to changes in moose density (Fig. 3). CONSTRAINTS OF PRED.4TOR AND The combination of the functional and PREY numerical responses represents the "total response." The total response may cause the Evolution has placed constraints of predation rate to be density dependent at both predatory and prey species, with low prey density and inversely density obvious implications for the relationships dependent at high prey density (Holling between them. In general, comparative 1959, Messier 1994). In a compilation of body size, strength, speed, and agility studies, Messier (1994) illustrated the total dictate a predator's ability to kill particular response of wolves to changing moose prey, while similar constraints on prey density. define which predators pose a threat to them. For example, predation by swift Other terms commonly used when foxes (V~rlpesvelo,~) on adult pronghorn describing predator-prey relationships are antelope (A~ltilocnprcramerica~ln) is highly "compensatory" and "additive" mortality. improbable, even though both species Ballard et al. (2001) defined additive occupy the same prairie . Similarly, mortality as occurring when the "additional the body size and defensive capabilities of risk of death does not cause reductions in (Microtzrs spp.) are no match for the other forms of mortality, but rather increases size and agility of coyotes, hence voles must overall ." On the other hand, rely upon other survival strategies. Such for compensatory mortality, the "additional physical and behavioral characteristics, or risk of death causes a reduction in other constraints, have developed over extensive forms of mortality so that overall mortality periods and represent an "evolutionary race" either does not change or is less than it between predator and prey. Some physical would be if additive." Kunkel and Pletscher abilities, e.g., the speed of pronghorn (1999) suggested that predation on cewids antelope, may even represent residual by several predatory species (mainly wolf developments from interactions with and , Puma concolor) was additive in predators now extinct (Byers 1997). northwestern Montana. Two terms also worthy of definition are "obligate" and HUNTING STRATEGIES "facultative" predator. An obligate predator is one that specializes on one primary prey The process within which predators species. Hence changes in the levels of the seek and kill has important implications to primary prey will generally influence a their impact upon prey. In the case of numerical change in the obligate predator. obligate predators, relief from predatory In contrast, a facultative predator is a dietary pressures on the prey will certainly occur generalist that switches among prey species with the numerical decline in predators and is thus buffered by changes in when prey become too scarce to support the abundance of any one prey species. A predators. Among mammalian systems, this facultative predator in a multi-prey system is perhaps best exemplified by the patterns can limit one prey species to low levels in abundance of (Lyt1.r cntzcriiensis) and because other prey maintain the predator snowshoe hare, or black-footed ferrets population. (ibf~rsiela nigripes) and prairie (Ci.rio~g,s spp.). \Vhen facultative predarors. ~viththeir abilities to switch from one prey source to another or even from directly to the number of coyotes and predacious to omnivorous diets, are inversely to the number of jackrabbits (the involved, relief from predatory pressures -jackrabbit ratio). If these scenarios may not be forthcoming. When a prey are correct, a transition from a Type I to a species comprises incidental portions of a Type I1 functional response occurs as predator's diet, the killing rate may be a jackrabbits mature. In the relatively simple matter of random encounter between ecological situation studied by Knowlton predator and prey and be strictly a product and Stoddart (1992), there was an apparent of the numerical abundances of both feed-back mechanism with the numerical predator and prey (Hollings' Type I abundance of coyotes dictated by the functional response). This type of situation abundance of their principle prey, adult can be particularly hazardous to prey species jackrabbits. Thus they propose predatory that are scarce. mechanisms that might partially explain the cyclic nature of some predator-prey As the relative importance of a prey interactions. species increases in a predator's diet, or the effort needed to acquire the prey increases, In addition to a component of strict the functional response assumes numerical abundance, the effect that characteristics of a Type I1 hnctional facultative predators, which switch from one response, with satiation placing an upper prey type to another, have on prey species limit upon the killing rate by individual probably reflects a complex integration of predators (for this discussion, we will ignore the relative abundance, efforts to capture, events like surplus killing, food caching, and the quantity and quality of reward etc.). In this case, the killing rate of the prey associated with each prey species. is directly proportional to the number of Consequently, understanding the role of predators but inversely proportional to the predation upon a prey species in this abundance of prey. This is perhaps best situation requires an understanding of the exemplified by iOlowlton and Stoddart's contexi within which it occurs. We (1992) hypothesis that predation upon acknowledge there also are components black-tailed jackrabbit nestlings by coyotes associated with habits, learning, and may be a Type I functional response, being traditions, among individual predators, but merely a matter of chance encounter those issues are beyond the scope we wish because the frequency and size of reward is to present here. insufficient for coyotes to actively hunt for them. On the other hand, they suggest that FACTORS INFLUENCIKG PREY predation upon adult jackrabbits is more POPULATIONS likely a Type I1 functional response, with coyotes actively hunting jackrabbits as a The role predation plays in wildlife dietary staple, but because it requires population dynamics follows many of the significant effort to capture adult constnlcts and theories established by earlier jackrabbits, they are only hunted when researchers working on insects. However, coyotes are hun,~. In this case, satiation of as in the entomological debates of decades the coyotes places an upper limit upon the past, the role of predation in the population killin2 of adult jackrabbits. Hence the d!namics of wildlife. particularly ungulates, fraction of adult jackrabbits killed is related is far from clear (Sinclair 1991. Skogland 1991, Messier 1991, 1994; Boutin 1992, (Sinclair 1979). Once rinderpest was Dale et al. 1994, Van Ballenberghe and eradicated, the wildebeest population tripled Ballard 1994). Much of the conhsion arises in size from 1963 to 1974 (Sinclair 1979). because predation is only one of many among ungulate species also factors influencing prey populations. may influence population levels. A theory Growth rates (increasing or decreasing) of currently proposed for the mule deer (0. prey populations may be affected by habitat hemionris) decline in some areas of the changes, severe weather (e.g., deep snow), western United States is competition with starvation, diseases, predation, human elk (Cervlu elaphus) and white-tailed deer. hunting, competition with other ungulates Equally disconcerting is the likelihood of (native and domestic), changes in sex and hybridization between mule and white-tailed age structure, as well as interacting deer, particularly in areas where habitat combinations of these factors (Ballard et al. modification increases the probability of 2001). For example, density-dependent interspecific hybridization (Hornbeck and food limitation and density-independent Mahoney 2000). Competition with adverse weather have been implicated as also has been implicated as a result factors regulating the numbers of in of cattle removing winter forage for mule the arctic tundra (Skogland 1985, 1990). deer and elk in the western United States. Gates et al. (1986) concluded that food Increased has resulted in loss limitation and snow conditions regulated of suitable habitat, especially wintering barren-ground caribou (R. t. groenlandicus) ranges, for many ungulate populations, in northern Canada. Mech et al. (1987) and although land conversion to agculture may McRoberts et al. (1995) reported the benefit some white-tailed deer populations. cumulative effect of snow depth over 3 Thus, while many factors affect prey winters influenced population parameters of population levels, for purposes of this paper, moose on Isle Royale and white-tailed deer let us focus on the effect of predation. (Odocoi!ezts virginianz~~)in Minnesota. In a counterpoint, Messier (1991) postulated FACTORS INFLUENCING that competition for food, not snow depth, PREDATION had a regulatory effect on moose, and that deer density and deer population growth was Skogland (1991) identified 9 factors inversely related to wolf density; with snow that may influence predation: habitat depth not a significant factor. Given the heterogeneity, prey refugia, nomadism same information, various researchers (temporallspatial availability of prey), buffer provide differing interpretations of the data. zones for prey, synchrony of the birthing It seems unlikely the debate over the season and aggregation at birthing, prey size hierarchy of factors influencing ungulate and age vulnerability, availability of population dynamics will soon be resolved. alternate prey, the ratio between the dominant prey species and alternative Disease is another factor that may (buffer) prey, and the effects of regulate some ungulate populations. The compensatory causes of mortality and the introduction of rinderpest to the Serengeti effects of alternative predator species. plains by domestic cattle caused high Several ofthese are related and interactions mortality amons wildebeest (Connochaetes amon2 factors may cloiid our understanding tnuril~ils) and buffalo (3.ncenis ccfler) of predator-prey systems Food is often a limiting factor on ungulate populations Seip (1992) suggests that one caribou (Sinclair 1979). Heterogeneity of habitat population was slowly increasing because has been proposed to influence predation on its migratory behavior kept the caribou prey populations. With increasing human separated from wolves and moose modification of the , prey throughout the year resulting in low wolf populations become fra,mented, or isolated, predation. Fryxell et al. (1988) postulated and more vulnerable to predators. Habitat that migratory ungulates on the Serengeti degradation may increase predation of may escape predatory regulation by their (Mustela vison) on water voles (Arvicola movements, while resident ungulate terrestris) in England (Barreto et al. 1999). populations might be more vulnerable to the Increased predation risk from habitat effects of predators. However, the seasonal fragmentation has been implicated as cause migratory patterns observed for ungulates for decline of game bird populations. In on the Serengeti are more likely due to contrast, modification to urban changes in forage quality across the may favor some prey species (i.e., white- landscape (Fryxell 1995), than predator tailed deer), yet dissuade large avoidance. Another antipredator strategy from these areas, forming a refuge from among ungulate species may result from predatory pressures. In a more natural reproductive synchrony and aggregation setting, Murie (1944) showed that Dall during and following the birthing process. (Ovis dalli) escaped wolfpredation by Reproductive synchrony and aggregation using steep terrain and cliffs as refugia. during birthing can flood territorial Similarly, Ferguson et al. (1988) suggested predators to the point that only a small that one population of woodland caribou portion of the reproductive effort falls prey reduced predation risk from wolves by to predators (Estes 1976). Although, residingon small islands. birthing synchrony is generally related to environmental seasonality and the plant The temporal and spatial availability growing season (Rutberg 1987). Perhaps an of prey also influences predator-prey equally importani effect may result from the relationships. In northeastern Minnesota, territorial nature of most carnivores, which white-tailed deer use buffer zones between limits their ability to respond numerically to wolf packs, which wolves avoid for fear of aggregations of prey, especially during intraspecific strife with neighboring packs periods of heightened vulnerability, as in the (Mech 1977). Nelson and Mech (1981) case of yarding among white-tailed deer or suggest that wolf predation regulates deer winter concentrations of black-tailed numbers, but the buffer zones between wolf jackrabbits (Smith 1987). territories allow sufficient numbers of deer to survive and these deer can reoccupy wolf Prey vulnerability is regarded as a territories when wolf numbers are low. major factor in predator-prey interactions. Subsequently, Nelson and Mech (1986~) Most predator-prey studies document how reported that the effects of snow depth and predators target young and old animals, vulnerability was the main factor regulating individuals in poor nutritional condition, or deer numbers, rather than wolf predation. prey that are weakened by disease or Migatory behavior is another mechanism physical abnormalities. In a classic study in that reduces the effect of predation. In the Alaska. ,Murie (1944) rsported that wolves LVells Gray caribou of British Columbia. killed Dall sheep that Ivere weak. diseased, or very old. Mech (1966) reported that which could exacerbate the interaction moose with heavy infestations of hydatid between a predator and another prey. In cysts (Echinococcus grand~rlosus),calves, Montana, Hamlin et al. (1984) documented and very old individuals were at greatest coyote predation as a major cause of mule risk of wolf predation on Isle Royale. In deer fawn mortality. Hotvever, when northeastern Minnesota, wolves killed a microtine rodents were abundant, mule deer preponderance of fawns, old male deer, and fawn mortality was low. They further individuals with abnormalities (Mech and concluded that vegetative production and Frenzel1971, Mech and Kams 1977, Nelson winter snow cover may regulate microtine and Mech 19866). Even the nutritional abundance, and thus fawn mortality rates. condition of the mother and grandmother Kunkel and Pletscher (1999) reported that may influence the vulnerability of first and where deer were present, the wolf-caused second generation fawns to wolf predation mortality rate on moose was lower than in (Mech et al. 1991). In Yellowstone areas where deer were absent. In contrast, National Park, Gese and Grothe (1995) Fuller (1990) believed the effect of wolves reported that adult elk killed by coyotes in on deer was exacerbated by the abundance winter were in poor nutritional condition, of moose in north-central Minnesota. based upon femur marrow fat indices. Stoddart et al. (2001) reported that coyotes Studies also have documented the high responded numerically as black-tailed vulnerability of new born fawns and calves jackrabbits increased during their 10-11 year to predation by a suite of predators (e.g., cycle. When the jackrabbit population Cook et al. 1971, Barrett 1984, Ballard et al. began to decline, coyotes switched to 1999). Vulnerability of a particular age domestic sheep and predation rates on lambs group in the prey population can influence escalated. population dynamics. Predators may remove a high proportion of neonates The effects of alternative predators on annual!y which may or may not affect a prey population can be substantial. In population levels (Linnell et al. 1995), or if Alaska, Gasaway et al. (1992) identified predators remove a high portion of the wolf and bear (Urstrs arctos) predation as a reproductive cohort (e.g., prime-age does), major factor limiting moose at low densities. the repercussions to the prey population may This multi predator system, with moose as be substantial. the primary prey, held the moose population well below carrying capacity. Kunkel and Availability of alternate prey can Pletscher (1999) reported that with wolf influence predator-prey interactions by recolonization in northwest Montana, the either diluting or exacerbating the effects of full compliment of predators (wolves, bears, a predator on their primary prey (Kunkel and ) brought about changes in the and Pletscher 1999). Dilution could be abundance of some ungulate species. They expected when alternate prey becomes more postulated that the mortality rate by all vulnerable than the primary prey (Carbyn predators on the cen~idpopulations (elk, 1983, Potvin et al. 1988); in which case the deer, and moose) was additive. 'new' prey may become the primary prey, buffering the former primary prey. In contrast, the abundance of one prey may cause a numerical response in the predator. Previous secrions identified many factors that may influence prey populations "predator pit," which refers to the "narrow and uredation rates. While these mizht- add band of densities between upper and lower confusion to predator-prey interactions, equilibrium points where ungulates can not scientists have developed several theoretical increase because of density-dependent models that allow testins of specific predation" (Ballard et al. 2001). The hypotheses and predictions among multiple stable states model may exist in competing models (e.g., Messier 1994). multi-predator and multi-prey systems. These models, examine the role of predation in ungulate population dynamics, are varied. Under the stable-limit cycle model, Four models widely used in predator- prey populations may exhibit regular cycles ungulate dynamics include: low-density of 30-40 years duration (Ballard et al. 2001). equilibria, multiple stable states, stable-limit Ballard et al. (2001) reported that severe cycles, and recurrent fluctuations (Boutin climate may influence the viability of young 1992, Van Ballenberghe and Ballard 1991, and the survival rates of young and adults. Ballard and Van Ballenberghe 1997, Ballard "Predation is density independent during et al. 2001). Similar models are presented population increases and inversely density- by Messier (1994) and Sinclair and Arcese dependent during population declines" (1995). Under the low-density equilibria (Ballard et al. 2001). Forage, climate, and model, prey populations are regulated at low prey density all interact to regulate the prey densities for long periods. The prey population. The stable-limit cycle model population remains at a low density until typically exists in single predator and single either a natural ohenomena or a decline in prey systems. predator abundance (e.g., predator control) allows the population- - to grow.- Limitation Sometimes a prey population by food is not important because prey fluctuates and never reaches a state of density never reaches carrying capacity. equilibrium. Prey densities change in When predators recover from low numbers, response to changes in climate, forage the prey population retums to a low density quality and quantity, and human harvest, but (Ballard et al. 2001). This model generally the primary factor most often limiting prey persists in systems with multiple species of density is predation (Ballard et al. 2001). At predators and prey where predators subsist high prey density, predation is inversely primarily on other prey. density-dependent. Prey may escape the regulatory effect of predation and attain a Under the multiple stable states model, higher density where ultimately food a prey population is regulated by density- competition causes a . dependent predation at low prey density Inversely density-dependent predation may until either a nahlral phenomena or predator accelerate or prolong the decline of the prey removal reduces the predator population, population. Perturbations cause the prey allowing the prey population to reach population to fluctuate without attaining a carrying capacity and become regulated by predictable density (Ballard et al. 2001). competition for food (Ballard et al. 2001). Both multi-predator and multi-prey systems Food competition then regulates the prey and single predator-single prey systems may pop~~lationat this higher equilibria e\.en exhibit recurrent fluctuations. after ths predator popillation retums to its forms; level. This model led to the tsm. It is unlikely that on? of th%s models will describe any predator-prey system at all At the elevated population level, forage times. Habitat conditions, human became suboptimal within the exclosure and populations, climatic events, and other the general health of the deer declined. factors are in constant flux. The acceptance Parasite loads increased, deer conceived of one model, without periodically later, bucks retained velvet longer, males reexamining the data on the entire system, shed antlers later, and goss reproductive would be foolish in light of the competing performance decreased (fie et al. 1979, ECle variables that influence both predators and and White 1985, Teer et al. 1991). prey. The relative merits of each model Eventually, the population declined to levels continues to be a source of debate within the comparable to outside the exclosure. scientific profession. Only through Compensatory mortality occurred with informative discussion, exchange of ideas, higher mortality among fawns 6-12 months developing data sets, and testing of of age, rather than the mortality occurring hypotheses will the debate prove fruitful. among post-natal fawns (Knowlton and Stoddart 1992). Essentially, the addition of EFFECT OF PREDATOR CONTROL animals above canying capacity required ON PREY POPULATIONS management action (e.g., increased harvest), or as in this case, compensatory mechanisms Predator control can enhance prey (i.e., malnutrition and ) returned populations if prey is at low densities the deer population to levels as before relative to carrying capacity. In Alaska, predator removal, but in a less healthy predator removal programs brought about condition (Knowlton and Stoddart 1992). irruptions of moose, which allowed for increased human harvest of moose If predator control is considered for (Gasaway et al. 1983, 1992, Ballard et al. enhancing ungulate populations, several 1991). In British Columbia, following factors should be considered. In a study in reduction of wolf numbers, was Quebec, wolf reduction was conducted on 2 enhanced 2-5 times for 4 ungulate species experimental areas while wolves were not and all populations increased (Bergerud and reduced on 2 control areas (Potvin et al. Elliott 1998). Similarly, deer populations in 1992). They found that deer populations south Texas increased following an increased in all 4 areas as a consequence of intensive coyote removal program (Beasom mild winters and recommended that wolf 1974). Predator control may obtain an reduction was not a viable management tool increase in ungulate numbers, but the in this context. Thus it is important that addition of animals can have consequences managers determine whether or not not often anticipated. In a study conducted predation is a limiting factor. Also, is the on the Welder Wildlife Refuge in south ungulate population below forage carrying Texas, coyote predation on white-tailed deer capacity? Considerations of scale, timing, fawns was substantial. To test if coyote and method of removal need to be addressed predation was a factor limiting population (e.g., what size of area is needed, and can growth, a 391-hectare exclosure was erected control be cost effective). As demonstrated on the site and the coyote density reduced throughout this paper, managers also need to inside. Deer densities in the exclosure consider ~vhatother factors may be limiting tripled compared to densities outside the or influencing the ungulate population. In a exclosure. and remainsd stable for 2-3 years recent rsvieiv of the relationships behveen predators and mule and black-tailed (0,h. in the predatory equation because they columbian~ts)deer, Ballard et al. (2001) contribute to the base determining the concluded that (a) the relationship of a prey numerical abundance of predators as well as population to forage carrying capacity was provide buffers for the prey and stability for critical to the impacts of predation, (b) prey the predators. The balance between prey populations do not respond to predator abundance and the upon which removal if prey is at or near carrying they depend is another integral part of capacity, and (c) when prey populations are understanding the interactions because as limited by predators and are far below their resources (i.e. food, cover, etc.) carrying capacity, predator removal could become scarce, their vulnerability to enhance prey survival, but increased hunter predation typically increases. harvest may be uncertain. Equally important is whether or not clear alternate One of the glaring lapses is an absence values or objectives of the prey population of long-term data sets with simultaneous are served. measures of the abundances and demographic parameters of predators and WHITHER FROM HERE? prey within individual . Such data sets provide the insights needed to Assessing the impact of predation generate the testable hypotheses that will upon prey populations is one of the more help define predator-prey interactions. daunting tasks facing the wildlife Ideally these data sets should include not profession. If it were easy, we would only routine measures of the abundances of already know the answers. Predation predatory and prey species of primary involves events towards the top of interest, but also those of alternate predatory ecological trophic schemes, with events at and prey species, as well as climatic lower trophic levels having repercussions conditions (especially deviant events), manifested in higher levels. Consequently, primary , and cause-specific attempting to unravel relationships at the top mortality among ine age classes of each without accounting for those below becomes prey species. Unfortunately, the difficulty illogical. However, we would be remiss of establishing and maintaining the interest without identifying potential means of and stream required for such improving our understanding. endeavors precludes many of us from such pursuits. It is indeed sobering to think that It now seems largely folly to attempt a for some long-term fluctuations (e.g., the comprehensive understanding of the role cyclic pattern in jackrabbit abundance in the and impact of predation on prey populations intermountain\vest), it may take 10-20 years independent of other ecolo~ical of data before we can generate the considerations. This may be less important appropriate questions, yet alone provide the in the case of obligate predators subsisting answers to them. on relatively few prey types but it becomes increasingly important as the number and We must recognize that by managing type of suitable prey and predators increases predation, \ye may be merely sustaining the complexity of the system (Table I). prey populations in that are ..\vailability and abundance of altsmate nizrsinal for other reasoris. In our quest for suitable prey constitutes a sipificani tern1 numbers. \ye ma)- bs m3kinz choices between smaller but robust and healthy prey numbers in the Thames catchment. Pages populations versus a more abundant but 31 1-327 in B. Shenvood and R. Bailey, perhaps less thnfty ones. Ultimately, editors. United Kingdom flood plains. The managers also must address the question of Linnean Society of London, London. whether efforts to manipulate predation will result in extending the life of prey long Barrett, M. 1984. Movements, habitat use, enough to reap some alternate value from and predation on pronghorn fawns in those animals. In doing so, we also need to Alberta. Journal of recognize that 'alternate values' are more 48:542-550. inclusive than harvest, and includes viewing, photographing, as well as simply Beasom, S. L. 1974. Relationships between preserving life forms for future generations. predator removal and white-tailed deer net productivity. Journal of Wildlife LITERATURE CITED Management 38:854-859.

Ballard, W. B., D. Lutz, T. W. Keegan, L. Bergend, A. T., and J. B. Snider. 1988. H. Carpenter, and J. C. deVos, Jr. 2001. Predation in the dynamics of moose Deer-predator relationships: a review of populations: a reply. Joumal of Wildlife recent North American studies with Management 52559-564. emphasis on mule and black-tailed deer. Wildlife Society Bulletin 29:(in press). Bergemd, A. T., and J. P. Elliott. 1998. Wolf predation in a multiple-ungulate Ballard, W. B., H. A. Whitelaw, S. J. system in northern British Columbia. Young, R. A. Jenkins, and G. J. Forbes. Canadian Journal of Zoology 76:1551-1569. 1999. Predation and survival of white-tailed deer fawns in north-central New Bmnswick. Bergemd, A. T., W. Wyett, and B. Snider. Journal of '?Jildlife Management 63:574- 1983. The role ofwolf predation in limiting 579. a moose population. Journal of Wildlife Management 47:977-988. Ballard, W. B., J. S. Whitman, and D. J. Reed. 1991. Population dynamics ofmoose Boutin. S. 1992. Predation and moose in south-central Alaska. Wildlife population dynamics: critique. Journal of Monographs No. 114. Wildlife Management 56:116-127.

Ballard, W. B., and V. Van Ballenberghe. Byers, J. A. 1997. American pronghorn: 1997. Predatoriprey relationships. Pages social adaptations and the ghosts of 247-273 in A. W. Franzmann and C. C. predators past. University of Chicago Press, Schwartz, editors. and Chicago. management of the North American moose. Smithsonian Institution Press, Washington, Carbyn, L. N. 1983. Wolf predation on elk D.C. in Riding Mountain National Park. Journal of Wildlife Management 47:963-976. Barreto, G. R., D. W. Macdonald, and R. Strachan. 1999. The tightrope hypothesis: Connolly. G. E. 1978. Predators and an explanation for plummeting Lvater predator control. Pages 369.394 in J. L. Schmidt and D. L. Gilbert, editors. Big declining white-tailed deer population in game of North America. Stackpole Books, north-central Minnesota. Wildlife Hamsburg, Pennsylvania. Monographs No. 1 10.

Cook, R. S., M. White, D. 0. Trainer, and Gasaway, W. C., R. 0. Stephenson, J. L. W. C. Glazener. 1971. Mortality of young Davis, P. E. K. Shepard, and 0. E. Burris. white-tailed deer fawns in south Texas. 1983. Interrelationships of wolves, prey, Journal of Wildlife Management 35:47-56. and man in interior Alaska. Wildlife Monographs No. 84. Dale, B. W., L. G. Adams, and R. T. Bowyer. 1994. Functional response of Gasaway, W. C., R. D. Boertje, D. V. wolves preying on barren-ground caribou in Grangaard, D. G. Kelleyhouse, R. 0. a multiple-prey . Journal of Stephenson, and D. G. Larsen. 1992. The Animal Ecology 63544-652. role of predation in limiting moose at low densities in Alaska and Yukon and Erlinge, S., G. Gbransson, G. Hogstedt, G. implications for conservation. Wildlife Jansson, 0.Liberg, J. Loman, I. N. Nilsson, Monographs 120. T. von Schantz, and M. Sylvkn. 1984. Can vertebrate predators regulate their prey? Gates, C. C., J. Adarnczewski, and R. American Naturalist 23:125-133. Mulders. 1986. Population dynamics, winter ecology and social organization of Estes, R. D. 1976. The significance of Coats Island caribou. Arctic 39:216-222. breeding synchrony in the wildebeest. East African Wildlife Journal 14:135-152. Gese, E. M., and S. Grothe. 1995. Analysis of coyote predation on deer and elk during Ferguson, S. H., A. T. Bergerud, and R. winter in Yellowstone National Park, Ferguson. 1988. Predation risk and habitat Wyoming. American Midland Naturalist selection in the persistence of a remnant 133:36-43. caribou population. Oecologia (Berlin) 76:236-245. Hamlin, K. L., S. J. Riley, D. Pyrah, A. R. W. Dood, and R. J. Mackie. 1984. Fryxell, J. M. 1995. Aggregation and Relationships among mule deer fawn migration by ungulates in relation to mortality, coyotes, and alternate prey resources and predators. Pages 257-273 in species. Journal of Wildlife Management A. R. E. Sinclair and P. Arcese, editors. 48:489-499. Serengeti 11: dynamics, management, and conservation of an ecosystem. University of Holling, C. S. 1959. The components of Chicago Press, Chicago. predation as revealed by a study of small mammal predation of the European pine Fryxell, J. M., J. Greever, and A. R. E. sawfly. Canadian Entomology 91 :293-320. Sinclair. 1988. Why are migratory ungulates so abundant? American Hornbeck, G. E., and J. M.Mahoney. 2000. Naturalist 131:781-798. Introgessive hybnd~zat~onof mule deer and ~vhite-taileddzer in southwestem Alberta. Fuller. T. K. 1990. Dynamics of 2 \Vildilfz Society Bulletin 28:IOiZ-101 j. Kidd, N. A. C., and G. B. Lewis. 1987. McRoberts, R. E., L. D. Mech, and R. 0. Can vertebrate predators regulate their prey? Peterson. 1995. The cumulative effect of A reply. American Naturalist 130:44S-453. consecutive winters' snow depth on moose and deer populations: a defense. Journal of Kie, J. G., and M. White. 1985. Population Animal Ecology 64:13 1-135. dynamics of white-tailed deer (Odocoilelrs virginianus) on the Welder Wildlife Refuge, Mech, L. D. 1966. The wolves of Isle Texas. Southwestern Naturalist 30:lOj-118. Royale. U.S. National Park Service Fauna Series No. 7. Kie, J. G., M. White, and F. F. Knowlton. 1979. Effects of coyote predation on Mech, L. D. 1977. Wolf-pack buffer zones population dynamics of white-tailed deer. as prey reservoirs. Science 198:320-321. Pages 65-82 ir~D. L. Drawe, editor. Proceedings of the First Welder Wildlife Mech, L. D., and L. D. Frenzel, Jr. (editors). Foundation Symposium. Welder Wildlife 1971. Ecological studies of the timber wolf Foundation Contribution B-7, Sinton, in northeastern Minnesota. USDA Forest Texas. Service Research Paper NC-51, North Central Forest Experiment Station, St. Paul, Knowlton, F. F., and L. C. Stoddart. 1992. Minnesota. Some observations from two coyote-prey studies. Pages 101-121 in A. H. Boer, Mech, L. D., and P. D. Karns. 1977. Role editor. Ecology and management of the of the wolf in a deer decline in the Superior eastern coyote. University of New National Forest. USDA Forest Service Brunswick, Fredericton. Research Paper NC-148, North Central Forest Experiment Station, St. Paul, Kunkel, K. E., and D. H. Pletscher. 1999. Minnesota. Species-specific population dynamics of cervids in a multi predator ecosystem. Mech, L. D., M. E. Nelson, and R. E. Journal of Wildlife Management 63:1082- McRoberts. 1991. Effects of maternal and 1093. grandmaternal nutrition on deer mass and vulnerability to wolf predation. Journal of Linnell, J. D. C., R. Aanes, and R. Mammalogy 72:146-151. Andersen. 1995. Who killed Bambi? The role of predation in the neonatal mortality of Mech, L. D., R. E. McRoberts, R. 0. temperate ungulates. Wildlife Peterson, and R. E. Page. 1987. 1:209-223. Relationship of deer and moose populations to previous winters' snow. Journal of Lotka, A. J. 1925. Elements of physical Animal Ecology 56:615-627. biology. Williams and Watkins, Baltimore. Macdonald, D. W., G. M. Mace, and G. R. Messier, F. 1991. The significance of Barretto. 1999. The effects of predators on limiting and regulating factors on the fragmented prey populations: a case study demography of moose and ~vhite-taileddeer. for the consewation of endangered prey. Journal of Animal Ecology 60:377-393. Journal of Zoology (Lond.) 247:487-506. hlessie:. F. 1991. Unsulzte population models with predation: a case study with the and E. J. Hofer. 1997. Kumerical responses North American moose. Ecology 75:478- of coyotes and lynx to the snowshoe hare 488. cycle. Oikos 80:150-162.

Messier, F., and M. Crtte. 1985. Moose- Pech, R. P., A. R. E. Sinclair, A. E. wolf dynamics and the natural regulation of Newsome, and P. C. Catling. 1992. Limits moose populations. Oecologia (Berlin) to predator regulation of rabbits in Australia: 65:44-50. evidence from predator-removal experiments. Oecologia (Berlin) 89:102- Moms, R. F., W. F. Chesire, C. A. Miller, 112. andD. G. Mott. 1958. Numerical responses of avian and mammalian predators during a Peterson, R. O., J. D. Woolington, and T. N. gradation of the spruce budwoxm. Ecology Bailey. 1984. Wolves of the Kenai 39:487-494. Peninsula, Alaska. Wildlife Monographs No. 88. Murie, A. 1944. The wolves of Mount McKinley. U.S. National Park Service, Potvin, F., H. Jolicoeur, L. Breton, and R. Fauna Series No. 5. Lemieux. 1988. Wolf diet and prey selectivity during two periods for deer in Nelson, M. E., and L. D. Mech. 1981. Deer Quebec: decline versus expansion. social organization and wolf predation in Canadian Joumal of Zoology 66:1274-1279. northeastern Minnesota. Wildlife Monographs No. 77. Potvin, F., H. Jolicoeur, L. Breton, and R. Lemieux. 1992. Evaluation of an Nelson, M. E., and L. D. Mech. 1986a. experimental wolf reduction and its impact Relationship between snow depth and gray on deer in Papineau-Labelle Reserve, wolf predation on white-tailed deer. Journal Quebec. Canadian .Journal of Zoology of Wildlife Management 50:471-474. 70:1595-1603.

Nelson, M. E., and L. D. Mech. 1986b. Rutberg, A. T. 1987. Adaptive hypotheses Mortality of white-tailed deer in of birth synchrony in ruminants: an northeastern Minnesota. Journal of Wildlife interspecific test. American Naturalist Management 50:691-698. 130:692-710.

Newsome, A. E., I. Parer, and P. C. Catling. Seip, D. R. 1992. Factors limiting 1989. Prolonged prey suppression by woodland caribou populations and their carnivores - predator-removal experiments. interrelationships with wolves and moose in Oecologia (Berlin) 78:458-467. southeastern British Columbia. Canadian Journal of Zoology 70: 1494-1503. Nicholson, A. J., and V. A. Bailey. 1935. The balance of animal populations. Sinclair, A. R. E. 1979. The eruption of the Proceedings of the Zoological Society ruminants. Pages 82-103 in A. R. E. (Lond.) 3551-598. Sinclair and M. Norton-Gnffiths, editors. Serengeti: dynamics of an ecosystem. O'Donoghue, M., S. Boutin, C. J. Ejebs vniversity of Chicago Press. Chicago. Sinclair, A. R. E. 1989. Population Taylor, R. J. 1984. Predation. Chapman regulation in animals. Pages 197-241 in J. and Hall, New York. M. Cherrett, editor. Ecological concepts. Blackwell. Oxford. Teer, J. G., D.L. Drawe, T. L. Blankenship, W. F. Andelt, R. S. Cook, J. G. Kie, F. F. Sinclair, A. R. E. 1991. Science and the Knowlton, and M. White. 1991. Deer and practice of wildlife management. Journal of coyotes: the Welder experiments. Wildlife Management 55:767-773. Transactions of the North American Wildlife and Natural Resources Conference Sinclair, A. R. E., and P. Arcese. 1995. 56550-560. Population consequences of predator- sensitive : the Serengeti wildebeest. Thompson, I. D., and R. 0.Peterson. 1988. Ecology 76:882-891. Does wolf predation alone limit the moose population in Pukaskwa Park?: a comment. Sinclair, A. R. E., P. D. Olsen, and T. D. Journal of Wildlife Management 52:556- Redhead. 1990. Can predators regulate 559. small mammal populations? Evidence from house mouse outbreaks in Australia. Oikos Van Ballenberghe, V., and W. B. Ballard. 59:382-392. 1994. Limitation and regulation of moose populations: the role of predation. Canadian Skogland, T. 1985. The effects of density- Journal of Zoology 72:2071-2077. dependent resource limitation on the demography of wild reindeer. Journal of Volterra, V. 1926. Variations and Animal Ecology 54:359-374. fluctuations of the numbers of individuals in animal species living together. Reprinted Skogland, T. 1990. Density-dependence in 1931 in R. N. Chapman, Animal Ecology, a fluctuating wild reindeer : maternal McGraw Hi!!, New York. vs. offspring effects. Oecologia (Berlin) 84:442-450.

Skogland, T. 1991. What are the effects of predators on large ungulate populations? Oikos 61:401-411.

Smith, G. W. 1987. Mortality and movement within a black-tailed jackrabbit population. Ph.D. dissertation, Utah State University, Logan.

Stoddart, L. C., R. E. Griffiths, and F. F. Knowlton. 2001. Coyote responses to changing jackrabbit abundance affect sheep predation. Journal of Range Management 54:15-20. Table 1. Gradients of increasing complexity in predator-prey interactions. Simpler predator-prey interactions to More complex predator-prey interactions - -- - Single prey system Multiple prey systems Stationary prey Mobile prey Resident prey Transient (migratory) prey Obligate predator Facultative predator Single predator system Multiple predator system Figure 1. Types of functional responses of predators to increasing prey density: (Type I) predator kills a constant proportion of the prey population regardless of prey density; (Type 11) predation rate decreases as predator satiation sets an upper limit; (Type 111) predator kill rate lags at low prey density owing to low hunting efficiency or absence of search image (Holling 1959). Figure 2. The functional response of wolves to changing moose density. -filling rate (number of moose killed per wolf per 100 days) was related to moose density (numberikm2) with a hyperbolic, Michaelis-Menton equation (data from Table 2 in Messier 1994). 0 0.5 1 1.5 2 2.5 Moose Density (no. per sq. km)

Figure 3. The numerical response of wolves to changing moose density. Wolf density (nurnberi1,OOO km2 presented on a log,, scale) was related to moose density (numberh2) with a hyperbolic, Michaelis-Menton equation (data from Table 2 in Messier 1994).